Lab 6

Part 1: Data Wrangling

Task 1.0: Install and load required packages

install_and_load <- function(pkg) {
  if (!requireNamespace(pkg, quietly = TRUE)) {
    install.packages(pkg)
  }
  library(pkg, character.only = TRUE)
}

install_and_load("tidyverse")
install_and_load("ggplot2")
install_and_load("BiocManager")
install_and_load("clValid")
install_and_load("scatterplot3d")
install_and_load("e1071")
install_and_load("gridExtra")
install_and_load("caret")

# install limma from Bioconductor
BiocManager::install("limma")

library(limma)

Task 1.1: Data Acquisition

• Read the dataset CSV file into R.
• By this point you should have two dataframes one with expression data only and the second one with expression data + phenotypes.

full_dataframe <- read.csv("Brain_GSE50161.csv")
print(dim(full_dataframe))

## [1]   130 54677

sample_ids <- full_dataframe$samples
full_dataframe <- full_dataframe %>% select(-samples)
rownames(full_dataframe) <- sample_ids

expression.data <- full_dataframe %>% select(-type)

Task 1.2: PCA Before QC

• Remove NAs by filling them with the means of their respective genes.
• Perform PCA on the imputed data.

gene_means <- colMeans(expression.data, na.rm = TRUE)

expr_imputed <- as.data.frame(
  Map(function(col, m) ifelse(is.na(col), m, col),
      expression.data,
      gene_means)
)

pca_res <- prcomp(expr_imputed, center = TRUE, scale. = FALSE)
pcs <- as.data.frame(pca_res$x)

head(pcs)

##          PC1       PC2        PC3       PC4        PC5        PC6         PC7
## 1 -119.93240 24.827291 -10.094289 62.736207 -35.900332  21.742327 -44.4305862
## 2  -68.84101  5.001261  41.672421 16.250916  32.350386 -64.411493  44.3314873
## 3  -57.25181 49.371208   1.737344 -2.646363  27.757115   1.688111  18.3426467
## 4   46.49128 58.391208  46.214517 31.314746   3.109689  -5.567812  -0.6694227
## 5  -66.54843 40.078145  26.095697 13.651545 -12.288291 -12.117699   4.2830189
## 6 -101.31301 33.126327  14.090266 54.989242 -13.783524   1.213111 -28.4467665
##          PC8       PC9      PC10       PC11      PC12        PC13       PC14
## 1   2.578779 -33.04951   2.84858  -0.545314 21.721916  -0.4533251  16.225963
## 2  41.655497 -24.10353  17.57165 -15.188171 -9.466229   7.2231154  -9.462181
## 3 -33.842216  23.90596 -11.01189   7.874218  5.069082  -6.3591703 -38.882183
## 4  33.989638 -13.21527  41.04846  25.864821 15.727538  -0.4740046   7.933242
## 5 -34.207421 -14.21203 -12.90762  23.210891 18.726082 -11.4974716  13.081504
## 6  -6.328682 -20.03668  12.02804   7.924769 26.087239  17.8073629   2.354601
##         PC15       PC16        PC17        PC18       PC19        PC20
## 1 -17.733297  28.861092 -12.7379890 -0.04086854  5.2828635 -11.6802420
## 2  -2.899471  25.420079 -10.4125127  2.36901833 -9.7992806  29.9229049
## 3   5.933572  -6.350106 -10.3000867 12.01559370 -0.8050119  -4.2787188
## 4 -17.530470 -22.600162  26.9718050 16.17343965 22.0217208   0.2474391
## 5  -1.379229  19.229522  -0.7032287 16.82180366 29.6313830   7.2459199
## 6  -7.236260  16.658568   0.8177497 10.11497142  6.5483476  12.2956208
##         PC21      PC22        PC23         PC24       PC25        PC26
## 1  10.912207 -7.952277  12.0058398 -17.16277314  20.450362  -4.9870240
## 2  10.697658 -8.822624 -17.9688065  -9.21390255  -3.085605   0.7925117
## 3 -11.437524 10.399054   0.8955279  -4.56460734  -1.898755   0.4229886
## 4 -10.137517 -8.526321   3.2398733  11.67742519 -37.929939  27.7908511
## 5 -17.340688 24.357745   4.4613648  -0.01439121   5.950370 -19.7640405
## 6  -1.317917  2.943043   3.3928980  -2.56966093  22.491375 -10.7572144
##         PC27       PC28       PC29        PC30      PC31       PC32       PC33
## 1  2.9525241  -1.213378  14.492722  -0.3014649  7.570983 -12.319356  0.5174670
## 2 10.9637774  26.540315 -13.637973 -21.1355150 -9.502167  29.476719 -9.7028655
## 3  0.7810011   7.434822   4.758866  -8.5328561  6.430716  -5.175358  6.0537150
## 4 30.8505826 -30.107687 -14.846850  31.0260516 -8.503102   4.425759  4.6794467
## 5  3.0990702  -9.803361  25.638226  -4.2817025 -2.822556  25.325940 -0.3977699
## 6 10.9438014  16.904352 -13.190625  -9.5518601 11.663195 -24.359029 -3.8735481
##          PC34       PC35       PC36        PC37        PC38        PC39
## 1  -1.7150459   4.983598  -8.085452  3.54679555  -0.2305111  -0.2473667
## 2  11.6787762 -15.256134   8.690024  3.06978463   2.5618143 -12.3823686
## 3   0.8112779  13.076281   4.704832 -1.83591401  -6.5119976  -7.8544720
## 4  11.6951795  22.811177 -21.961723 -5.89540937 -29.1207871  -2.3355639
## 5 -10.9536018  13.679179   1.569094 -1.88690977  -0.2116142   5.7023866
## 6 -11.8380868  15.150467  14.683500 -0.07288528  -2.2302969  -7.9001986
##        PC40        PC41       PC42       PC43       PC44      PC45       PC46
## 1 -1.319043   2.1778335  6.5724859  -2.791390  -4.596974 0.9879256   2.653356
## 2 14.155200   3.5955975 15.5636197  36.035278  15.366230 8.5768259 -11.370283
## 3 -5.518397   5.8905983  4.7108949   4.573744   7.152647 1.2300964   8.615083
## 4  2.778788 -15.9044002  4.4174247  10.440855  -6.207317 7.1493639   8.292508
## 5 -9.214930   8.8924003  3.9321105   2.265603 -21.745650 7.9116596  -8.190611
## 6 -3.806003  -0.9667394 -0.3269828 -14.213204  10.567662 1.9600125  -1.553188
##         PC47       PC48      PC49      PC50      PC51       PC52      PC53
## 1   7.333299 -11.089065 -5.122409  6.794641  1.115385   3.714771  4.652493
## 2  14.752465  -5.461702 -6.189480  4.114286 -1.447320  23.316752 -4.207146
## 3   3.908798   6.346896 10.571855  6.875930 10.209163  -1.003060 -4.995462
## 4  36.059320   6.917615 25.660189  3.986815 -4.353559   4.089820  8.430732
## 5   7.221954   5.635017 -4.336607 -1.936774 -7.761778  26.836764 -8.596473
## 6 -11.004656 -16.250333 11.435577  9.415488 10.575381 -20.453578 -1.837568
##        PC54        PC55      PC56        PC57      PC58        PC59       PC60
## 1 -6.931218 -14.9459900  4.732695   3.0036562 11.802773 -10.4411937 -7.1170526
## 2 16.666096 -15.1694873 23.534413   2.9383622 -1.707837   6.4315584 -8.4720414
## 3  7.482779  -0.5941631  5.913367   2.1451077 12.571519  -0.8059244 -1.0685679
## 4  6.451074   5.4690214 -3.078068   7.1495762  9.723485   2.7593208 -0.9296478
## 5 -9.930130  18.8788271  2.352520 -17.1820306 -9.886098  -2.5609620 -5.4639143
## 6  8.377588  -3.3808112  3.930010   0.3704952 -2.964980   7.3372143 11.4637287
##         PC61       PC62       PC63       PC64       PC65        PC66       PC67
## 1 -7.9155437  -2.768513   2.829537  0.9441257  3.2308976   6.7475022  -3.474957
## 2 -6.4272015  -5.103617 -18.622389  2.1226011 -5.0474752 -17.1574691   2.757306
## 3  0.2180210   5.570324   3.336870 -0.8296533 -0.8045933  13.2733910   1.451469
## 4  0.3528927 -18.312729 -12.381985  9.6373256 14.8497166   4.9069475 -11.416655
## 5 -4.4105890  18.639486   2.490264  1.2014827  6.2657592   0.0240235 -17.529961
## 6 19.0370243 -13.444373 -11.183459  4.9240477  4.1540932  -3.6712308   2.030822
##         PC68       PC69        PC70          PC71        PC72        PC73
## 1 -2.2038827  -2.733830  -1.2751170 -9.474963e-04  -0.5237163  -0.3200528
## 2  3.5631105   2.551003 -10.6238110  7.159030e+00  -9.0068837  -7.7534084
## 3  8.3955400   3.900256   2.1981157 -1.801394e+01  -1.0995887  -7.1363924
## 4 -0.3001657 -12.087313   0.6596114 -8.226976e-02  -0.3636192   2.8838148
## 5 -4.8088808  -9.914981  16.5271941  9.649029e+00 -14.9673032 -11.8520264
## 6  6.6530404  -7.841669  -0.1268335  1.188596e+01  -8.6482684  -1.8330667
##          PC74        PC75       PC76       PC77        PC78       PC79
## 1  10.1075485  10.4409113 -0.0463334 -7.3931320  -1.8211545 -10.072533
## 2   5.9355256  11.8617796  2.1889176 -0.3343739  -5.2273931  -3.611012
## 3 -12.1127563  -0.3959444 -3.0851607  0.1018559  -0.4885251  -8.794861
## 4  -0.2392998  -1.5450462  1.6139016  3.1003416   3.7701116  -1.132516
## 5 -14.5914935  13.5943389  1.9564865  3.3731116   3.5846095   7.529591
## 6   1.4388618 -14.6384632 -0.8247056 24.5493133 -13.5089376   3.767613
##         PC80          PC81       PC82        PC83       PC84       PC85
## 1  5.5281570  1.182935e+00  1.7807670   4.5511577   4.882353  6.0323015
## 2  2.4977534  2.972756e-05 -9.4225982  -4.7374685   4.972373  2.7994594
## 3  6.3242225 -5.200839e+00 -7.4317382 -10.9155743  15.937461  0.9941390
## 4  0.9810064 -5.607016e-01 -0.4264722  -0.7917195  -5.136132 -0.2483665
## 5 -6.6024913  5.238729e+00 -2.7903805   1.8433221  -3.437882 16.1589037
## 6 -0.7416930  9.968909e+00  2.0326027 -11.8690452 -16.626941  7.6932400
##          PC86      PC87      PC88      PC89       PC90        PC91        PC92
## 1   5.4366908 -9.577054 -4.187302 -1.485353 -4.1969751  -0.7471969   3.3931635
## 2   3.1280555  5.467162  1.725334 -6.783418  5.4006017  -0.4370818  -1.3552201
## 3 -11.1341235 -2.179700  7.810434 -5.221097 -4.3290368   1.8671692 -10.4442715
## 4  -0.7272288  1.297104 -5.481196 -3.888267 -2.2063459   2.7671047   0.6790765
## 5  -1.4407917  3.983407  3.223037  9.290440  1.9256848 -14.7276216   7.9317249
## 6  -4.6010797 16.617697 10.125812  7.642596  0.5458442   8.8100281   7.1846428
##          PC93       PC94       PC95      PC96       PC97       PC98       PC99
## 1   7.9537661  8.3185606 -1.0884919 -6.084530 -4.5179499 -2.3318725 -0.2928333
## 2   1.3397646 -3.8699964  2.2712827 -1.579663 -0.6650955 -0.9139238 -1.2864487
## 3  -0.3896491  4.5749364 -4.7177100 13.274123 11.2456210 -0.1181166 -0.8673114
## 4   0.7452045 -0.5762996 -0.6796637 -2.576290  2.8622492 -5.8761905 -1.5787022
## 5  -5.2657071 -1.5060604  0.1501855  6.418179 -2.5047912  2.9048479 -8.6246823
## 6 -11.6966509 -5.6739051 -1.5541965  1.195087 -0.4288195 -8.3633413 -8.3099550
##        PC100     PC101     PC102      PC103      PC104        PC105       PC106
## 1  8.1551337  4.983589  4.367098  5.0928125  0.5700310 -13.56149509 -12.2257390
## 2 -0.5223751 -2.198926 -1.902898 -1.3692409  1.5186820  -1.99018323  -1.5209899
## 3 -3.4379898 15.544213 -4.447825 13.6391105 -0.7170527  -0.34275533   9.3055093
## 4  1.3485570 -1.351270 -1.600846  0.1045436  0.7060513  -1.81218692  -2.1266497
## 5  4.6617562 -3.282393 -7.127533  5.7689584  0.5200250   0.71604675   1.0090677
## 6  2.9606312  2.800001 -5.889269  0.1419995 -0.8079020   0.08423659   0.2147539
##          PC107       PC108      PC109      PC110      PC111       PC112
## 1 -1.439820711  7.96870163  1.5162638 12.2504257  5.8177077   2.7118322
## 2  1.266611394  0.09128964 -2.0992102  0.8977650 -0.2605935  -0.4867882
## 3 -7.203929936  1.61008096 -3.7070817 15.3320283  2.4284586 -21.1323940
## 4 -1.173660080  0.44212133 -0.4132916  0.5824686  0.7260252  -1.6454786
## 5  0.007905867 -1.20157071  1.7289882 -0.1562909 -0.7481727  -1.5029046
## 6 -1.532236537  0.99499234  3.1843217  4.3253168 -5.1056222   0.4203596
##        PC113      PC114      PC115      PC116       PC117       PC118
## 1 -2.0172946 -3.8757956 -1.7072924 -3.9758299 -10.4526737  1.10989511
## 2 -0.5068230  0.6483819  0.5343626 -0.4426099  -0.4057555  1.36815882
## 3 -0.9763932 -4.5221772  8.4676890  0.3584558   4.9334485  1.84564820
## 4 -0.9704757  1.3689293 -0.2868778  0.7076164  -0.2494771 -0.02014409
## 5 -1.9199575 -0.5164828  1.6630291  1.6076865  -1.7841272 -0.48374961
## 6 -0.4475982 -0.8253509 -0.4344208  4.6689466  -4.5984338  3.27295799
##         PC119      PC120      PC121      PC122      PC123      PC124
## 1 25.02944040 -6.0129004 -1.8884602 -6.8314603  8.8351460 -2.6999500
## 2 -0.68665331 -0.3593652  0.3784540  0.6600305 -0.5475772  0.1077035
## 3  2.06229396 -2.6615028 11.5892595  3.7224567 -1.3191123 -3.2861830
## 4 -0.12143832  0.1594981 -0.1661852  0.2632925 -0.2025322  0.9721007
## 5 -3.16841893 -0.8161031  0.1129896 -0.2165675 -0.4454278  1.8375385
## 6  0.05170174 -2.0146117 -2.0170764  1.3640659 -0.2329948  2.3670492
##         PC125      PC126       PC127       PC128       PC129         PC130
## 1 -2.97903256  3.7988662 -1.72371526  0.61597437 -0.58349585 -2.235561e-13
## 2 -0.31644179  0.1072522  0.20485887 -0.01968366 -0.30481594 -4.322571e-13
## 3  0.09220287  2.5044990  0.33828898 -0.76225701  1.18996028 -1.039375e-13
## 4  0.06192446 -0.2548097  0.50360056 -0.10426067  0.05544012 -1.494752e-14
## 5  0.29736806 -0.9379206  0.06013985 -0.35960285 -0.53202148 -4.139866e-13
## 6 -1.65334300  1.6733750  0.82040446  0.79877491 -0.15781011 -8.564705e-13

Task 1.3: PCA Before QC [Visualization]

• Visualize the PC1 vs PC2, PC1 vs PC3, and PC2 vs PC3 plots.
• Colors are according to the phenotype of each sample

pcs_plot <- pcs %>%
  mutate(phenotype= full_dataframe$type)

p1 <- ggplot(pcs_plot, aes(PC1, PC2, color = phenotype)) +
  geom_point() +
  ggtitle("PC1 vs PC2 Before QC")

p2 <- ggplot(pcs_plot, aes(PC1, PC3, color = phenotype)) +
  geom_point() +
  ggtitle("PC1 vs PC3 Before QC")

p3 <- ggplot(pcs_plot, aes(PC2, PC3, color = phenotype)) +
  geom_point() +
  ggtitle("PC2 vs PC3 Before QC")

grid.arrange(p1, p2, p3, ncol = 1)

Task 1.4: Data Cleaning

• Remove outliers from the data.
• Re-impute the data after removing outliers.
• Perform quantile normalization on the data (Normalizes expression intensities so that the intensities or log-ratios have similar distributions across a set of arrays).

# Replace outliers with NA using apply
expr_clean <- as.data.frame(
  apply(expr_imputed, 2, function(vals) {
    mu <- mean(vals)
    sdv <- sd(vals)
    outliers <- abs(vals - mu) > 3 * sdv
    vals[outliers] <- NA
    return(vals)
  })
)

# Re-impute missing after outlier removal
new_gene_means <- colMeans(expr_clean, na.rm = TRUE)
expr_clean <- as.data.frame(
  Map(function(col, m) ifelse(is.na(col), m, col),
      expr_clean,
      new_gene_means)
)
# Quantile normalization
expr_norm <- normalizeBetweenArrays(as.matrix(expr_clean), method = "quantile")
expr_norm <- as.data.frame(expr_norm)

# Explanation:
# `normalizeBetweenArrays()` adjusts the distributions of gene expression
# values across samples to be the same; the quantile method aligns
# empirical quantiles, ensuring comparability across arrays.

Task 1.5: Data Inspection

• Check the distribution of the data before and after normalization.

pca_norm <- prcomp(expr_norm, center = TRUE, scale. = FALSE)
pcs_norm <- as.data.frame(pca_norm$x)
head(pcs_norm)

##         PC1      PC2       PC3       PC4         PC5        PC6        PC7
## 1 -91.66181 18.46092 -37.83219 59.902227 -11.6308255  44.270490 -31.863292
## 2 -49.21615 12.86853  51.17655 10.759418  31.3222788 -21.248329   4.263852
## 3 -42.23751 36.59539  17.76164 -6.337331  15.3096276 -19.771474  11.278661
## 4  60.70806 18.52200  26.07607 27.492593  -0.2948381  14.241085  34.065638
## 5 -38.38705 31.18689  16.72598 31.990190  -1.1355782   5.522872 -43.355787
## 6 -72.47518 20.35510 -11.59330 56.283563  -3.3253633  20.706451 -19.413471
##          PC8        PC9        PC10       PC11       PC12       PC13       PC14
## 1 -26.269361  21.941904 -19.7820812  12.670441  -2.092730  9.6767388  -6.158493
## 2 -12.689859 -33.800869   8.1477639  16.664084 -39.748517 10.9538537 -20.321550
## 3  14.800202 -12.845988   0.1133992 -26.319503  26.577826 -5.3256252  -4.484167
## 4 -19.864696 -13.145974  -7.8721014  -3.101286 -23.257807 -3.1891688  10.728850
## 5   1.551413  -4.025139 -27.0317463   1.938334   9.762934 -9.8919824   2.229049
## 6 -17.846032   1.367711 -21.3627106  -2.224670  -4.354057  0.6385803 -19.009796
##        PC15        PC16        PC17       PC18       PC19      PC20        PC21
## 1  9.809573  -8.8303046   0.9372469  -2.192460 -12.072512 -9.230792   7.3139944
## 2  6.673205   3.8689931  -4.4599414  -2.288978  -4.645045  7.549552 -18.0559547
## 3 18.536527   4.5956411  14.5688468   5.710921  15.253255  8.803289  -2.1413740
## 4  4.643211 -24.2815524 -11.4683668   5.690325  18.379709 -7.703705  12.7356699
## 5 12.800297  11.6209587  -5.3659415  -5.654590   1.157146  1.108098  -1.3834650
## 6  1.536763   0.9796452   1.4619540 -11.115324   5.552319  8.453915  -0.5328217
##          PC22       PC23       PC24      PC25       PC26      PC27        PC28
## 1  -0.5746241   7.603374 -4.7906116  5.067100 -4.0906724  1.644949  -1.6336518
## 2   2.6872751   9.807959 -0.1542869 -3.157790 -5.2665953 15.940312  -9.6124656
## 3   5.4361476  -3.713468 -1.4347187  6.119747  0.3885904 -5.434759   2.0648362
## 4 -11.6234959   7.207767 -1.0721257 -4.065366  4.1228041 -9.054264  -0.7509123
## 5 -18.7680053 -20.080747 -9.2402099  9.074533 -2.1129410 10.557703   9.8034174
## 6  -6.0119313   8.954091 -1.8648268  5.415404 -9.7154049  3.319007 -20.5770884
##        PC29       PC30      PC31       PC32       PC33       PC34       PC35
## 1 12.114949  -7.178393 11.754220 -20.848068  10.922862  -3.156749   4.933811
## 2 -3.396890 -10.515732  3.493692  14.329714   5.607671  15.318675 -20.829849
## 3 13.493919  -1.697055 -2.242778  -2.385621   8.589111   9.137029  -5.182252
## 4  3.886464  -7.013062 -1.216654  17.371273 -17.466998  -7.729865   5.004284
## 5 10.357448  -4.222694 -1.605271  -1.837850 -13.379107 -14.342129 -10.450935
## 6 17.750246  13.291496 -9.371251  -2.617660  -5.606546   2.936529   6.719579
##         PC36        PC37       PC38       PC39      PC40        PC41
## 1  -6.413467  -5.2524333   3.231752  -7.992833 22.055167   0.8394198
## 2  -3.166676  -1.4460244 -22.511665 -22.120329 -9.630694  -3.9713727
## 3   0.091110 -10.7441199   6.934720   2.923016  7.930806  -7.6469090
## 4 -13.627912   8.1912480  26.045077   6.538160  8.181689   4.4509212
## 5 -15.222542   0.2663931  -6.204964  10.822000  1.120565  10.7676081
## 6  16.412011  -0.6620488   8.909985   4.486773 -1.078688 -10.4995101
##          PC42       PC43        PC44       PC45        PC46       PC47
## 1   1.0524177   2.440311  -4.0102734  -7.006467   4.5889260  3.5364301
## 2 -20.8753382 -13.259880  -3.0514579  -1.410135   7.8928190 10.5165599
## 3   0.2925945  -2.368455   5.9219017  -1.786099   5.8266783 -2.0406224
## 4  -9.7354491   1.846787 -21.4880085 -24.318228 -15.8862138 -0.4665691
## 5  -3.4349197   7.818752  -0.9950213   6.774443  13.4093216 11.9064071
## 6   3.3706001   1.900727  16.1459955   6.983747   0.7669582  0.5892932
##         PC48        PC49       PC50       PC51       PC52       PC53      PC54
## 1  -2.852321  -0.2971495 -10.315050  -3.784924   6.845927   6.364289 -7.261437
## 2  -6.907769  -7.7271240  -2.814964 -11.759129   9.548651  -7.875377 -5.109196
## 3  -3.838071  -8.9275420  -6.733576  -8.859117   4.451029  -8.778108 -3.764561
## 4 -12.083608 -15.3374698   8.521742  -2.162885  -6.412029 -20.584367 20.708489
## 5 -10.349380   5.3365768   2.386373 -10.209070  10.344846   8.791504 27.737784
## 6  10.776384  -0.1327187  -1.247270   6.542193 -12.675413   5.356517 -5.144122
##          PC55        PC56      PC57       PC58      PC59      PC60        PC61
## 1  -1.0788645  -0.1586225 -2.552087   4.412386  4.913328  2.797164 -5.53905449
## 2   0.9622349 -12.1406233 -9.033134 -15.544541  9.406763 18.485204 -0.03456041
## 3  -5.8612808   7.7045956  3.567098  -3.818155 -1.445936  1.012271 -1.83164592
## 4 -10.5405024  12.2377230 -1.694580  -1.002441  1.657736  6.233066 -3.72858290
## 5   4.7113017 -10.3222591  4.106768  -4.020470 -7.691563 -2.599476  5.47840610
## 6   7.1882505   1.3025077 13.865972  -4.412213 -6.354921  8.715718 -8.95176872
##         PC62        PC63       PC64      PC65       PC66      PC67       PC68
## 1 -3.8179418  -7.4646587   3.067144  8.085455  0.6388518  3.656808  0.8663305
## 2  3.3913750  -9.6177989 -10.192862 -8.691382 -5.5068859  4.811400  5.5521377
## 3  7.6284242  -2.1887704  13.590732 -9.823012 -2.0711729  2.164336  3.9797517
## 4 -0.9022489   0.9863942  20.825249  9.212996 -3.8567702 17.017686 -5.6955195
## 5 -1.4738247   1.4525447   2.909124  5.200892 -1.6196276  5.382067  3.7924116
## 6  7.1758605 -10.7893591  -6.468697  6.971455 -0.9986690  6.476056  4.6497677
##         PC69      PC70       PC71      PC72       PC73        PC74       PC75
## 1  -9.181489 -1.882496 -1.4634980 -4.887243  7.1642665   4.1532957  1.9117199
## 2  12.575587 -5.855922 12.4207768 -2.395147 -2.8526307   0.7214526 -1.6804544
## 3  -2.136191 -1.220240 -1.0916327 -1.163674  0.5821001   4.3741150  0.1818238
## 4  16.294937  6.759186  9.2668613  2.829066 -3.5552940  -0.1444241  8.8751617
## 5 -21.618329 20.083830 -0.9981809  4.935875  3.4716174 -12.1627118 -1.0972785
## 6  21.698613 -7.319587 -9.6244483  5.522868 -1.4485206   8.3341910 -5.8031339
##          PC76       PC77        PC78       PC79      PC80      PC81      PC82
## 1   0.7947095  5.0135957   0.8191000 -4.4957293 -1.043270 -4.754723 -3.865914
## 2 -12.9710302 14.6683528   0.3096906 -5.9740855 -3.249420  4.511326  5.996589
## 3  -4.2881612 -0.9794356   8.1461826  2.0792609 -2.901278 -5.160753 -5.118885
## 4  -1.0774081 10.6499650  -3.9600063 10.8750283 -1.319446 10.280089 -5.169247
## 5 -26.2091337 -2.2965439  -9.1602591 -0.5520696 -5.716144  0.655327 -3.280247
## 6  -4.5354008  0.6635965 -12.3873661  7.7938174  8.878827 -3.433235 -8.253099
##         PC83      PC84      PC85       PC86       PC87       PC88       PC89
## 1 -7.6629411  1.960395  4.138042  3.1020685   1.810274   6.909908   4.746096
## 2  0.4772508 10.616772  9.557230 -4.1426793  -2.057030  -3.502673   7.798675
## 3 -8.3509586  2.828436 -9.030920  1.6346302 -10.449166  -3.371392   1.096440
## 4 -4.9072414 -4.951317 -4.527349 -0.2642927   8.266561  -4.953660   1.985734
## 5 -1.3092021 -1.224719 -5.052514 -2.7674427  -1.364435 -10.492675  -2.606502
## 6 13.8789543  1.025341 -6.821520 -4.1431135  -1.847044  -3.397619 -16.289359
##        PC90       PC91       PC92       PC93      PC94        PC95       PC96
## 1 -1.529516 -5.2845135  -8.174418  -1.272759 -6.025442  13.4201725  0.8010566
## 2 11.518022 -3.5711755  -1.038111   5.984955  1.186786   5.0191625  8.1411682
## 3  1.091852 -5.6478300   2.798466   4.747722  5.080899 -18.4892043  5.1647237
## 4  5.004409  0.7899476   2.500173  -3.151771  1.158003  -1.2073756 -6.0846403
## 5 -2.686007 -5.2152120 -11.672044   4.107813 -6.406675  -0.4995136 13.3422127
## 6 10.711904  4.2592968  -5.429720 -13.832931 16.788286  -1.0956164  1.0130204
##        PC97       PC98        PC99     PC100     PC101      PC102     PC103
## 1  2.367454  5.5931108 -27.5287649 -7.645025 -9.148043 -6.7685248  5.273377
## 2  8.062327 -1.2312229   2.0866429 -6.019498  3.765601  0.7079325 -4.579788
## 3 -3.019974  0.2099759  -3.9638555  2.618486 -5.366366 -3.5413111 -7.685771
## 4  3.178207  3.3857624  -0.4912187 -5.542967  4.350661 -1.5407766  4.054482
## 5  2.540261 -1.3813150   7.4476240  8.399179  2.695332  8.6201208  9.770047
## 6  6.000768 -1.2660664   5.4977848 -2.018941  3.549158 13.7709002  6.897728
##       PC104      PC105     PC106       PC107       PC108     PC109       PC110
## 1 14.496731 19.6371211  6.356148 -12.2677041 -13.9200240 -3.595968 -1.65115338
## 2  1.357059 -2.4598604  6.951901   0.3510554   0.5239934 -6.882441 -1.32268411
## 3 -1.161830 -1.0442822 -6.045696   5.7219071   2.4664381  5.377590 16.24279971
## 4  2.505302  1.7493837  3.718209  -1.1230799   0.9009402 -1.459771  0.04032209
## 5  2.130557  1.9384227  1.674512   5.0021484   7.5244488 -2.433567  0.65793478
## 6  7.628291  0.7962132 -3.393428  -5.5935848   4.4472168 -2.877378  2.24273124
##        PC111      PC112       PC113     PC114       PC115      PC116
## 1 13.9580530  1.7369893  -2.9321822 -3.669606  -5.7891983  0.8156073
## 2  3.8482340  0.2296650   3.0586164 -1.513745   0.3377609  0.1259450
## 3  8.4837213  6.9916989 -15.8910143 -5.251490 -10.2821751 19.7485020
## 4  0.6570842 -0.3191966  -0.9492123  2.666985  -0.3211794 -1.8888274
## 5 -2.4077897  2.7641046   3.0072321  1.082876  -5.8200575 -1.0544731
## 6 -0.2214633  8.1470240  -1.1848480  1.552709 -13.4913986 -3.4431561
##         PC117      PC118        PC119      PC120      PC121      PC122
## 1   0.6616463  0.6133686  0.563560564  1.9085029 -2.3935672 -0.6850051
## 2  -4.3129499  2.4268223  0.888416753  0.9438208  0.2585674 -1.6764848
## 3 -15.4746522 -6.6602657 -1.236849828 -7.2611597 -6.0410503  3.8770047
## 4  -2.7697305 -0.3371165  0.333847172  0.5645988 -0.1977469 -0.7046990
## 5   0.2192694 -2.2597183 -0.009373023 -1.9448278  4.4689427 -2.0516717
## 6  -0.3196146  0.2633060  9.592031001  0.2493069  1.4193979  0.3756665
##        PC123      PC124      PC125      PC126       PC127       PC128
## 1  0.5424589  0.3837293 -1.5011999  1.8531203  -1.1769630  3.06986464
## 2  0.6896699  0.2181854  0.2175052 -0.4007371  -0.9276218 -1.16350050
## 3 -3.1537460  0.2285349  1.0800451 11.8957053 -13.3602421 -7.21790665
## 4  0.6694846  0.4613463  0.7999281 -0.4212578   0.2891144 -0.85893361
## 5 -0.5201540  3.0635632 -1.5077222 -1.4310458  -0.2967790 -0.01557093
## 6  0.4670280 -2.9016470  0.2483461 -2.8377712  -0.4806704 -0.16832687
##        PC129         PC130
## 1 -1.1893807  8.371119e-12
## 2  2.0214683 -2.908535e-13
## 3 -0.9407434 -1.314908e-13
## 4  1.2130060 -4.977003e-14
## 5 -0.2541687 -2.670978e-13
## 6  3.9132028 -1.527958e-12

• Visualize the PC1 vs PC2, PC1 vs PC3, and PC2 vs PC3 plots after normalization.
• Colors are according to the phenotype of each sample

pcs_norm_plot <- pcs_norm %>%
  mutate(phenotype= full_dataframe$type)

p1 <- ggplot(pcs_norm_plot, aes(PC1, PC2, color = phenotype)) +
  geom_point() +
  ggtitle("PC1 vs PC2 After QC")

p2 <- ggplot(pcs_norm_plot, aes(PC1, PC3, color = phenotype)) +
  geom_point() +
  ggtitle("PC1 vs PC3 After QC")

p3 <- ggplot(pcs_norm_plot, aes(PC2, PC3, color = phenotype)) +
  geom_point() +
  ggtitle("PC2 vs PC3 After QC")

grid.arrange(p1, p2, p3, ncol = 1)

Part 2: Analysis

Task 2.1: Regression Analysis

Read the list of top 5000 genes from the file top_5000.txt.

• Select only these genes from the normalized data.

# Encode phenotype: Tumor = 1, Normal = 0
phenotype_binary <- ifelse(full_dataframe$type == "normal", 0, 1)

# Load list of top 5000 genes
top_genes <- read.table("top_5000.txt", header = FALSE, stringsAsFactors = FALSE)

# Extract only these genes from normalized data
genes_subset <- expr_norm[, colnames(expr_norm) %in% top_genes$V1]

Perform logistic regression for each gene.

• Create a dataframe with the p-values and coefficients for each gene.

# Perform logistic regression for each gene
results_without_pc1 <- data.frame(Gene = character(),
                      p_value = numeric(),
                      coefficient = numeric())

for(gene in colnames(genes_subset)) {
  model <- glm(phenotype_binary ~ genes_subset[,gene], family = binomial)
  results_without_pc1 <- rbind(results_without_pc1, data.frame(
    Gene = gene,
    p_value = summary(model)$coefficients[2,4],
    coefficient = summary(model)$coefficients[2,1]
  ))
}

Perform logistic regression for each gene with PC1 as a covariate.

results_with_pc1 <- data.frame(Gene = character(),
                      p_value = numeric(),
                      coefficient = numeric())

for(gene in colnames(genes_subset)) {
  model <- glm(phenotype_binary ~ genes_subset[,gene] + pcs_norm$PC1, family = binomial)
  results_with_pc1 <- rbind(results_with_pc1, data.frame(
    Gene = gene,
    p_value = summary(model)$coefficients[2,4],
    coefficient = summary(model)$coefficients[2,1]
  ))
}

Filter the results to get significant genes (p < 0.05) and sort them by p-value.

# Get significant genes (p < 0.05)
significant_genes_no_pc1 <- results_without_pc1 %>% filter(p_value < 0.05) %>% arrange(p_value)
significant_genes_pc1 <- results_with_pc1 %>% filter(p_value < 0.05) %>% arrange(p_value)

Task 2.2: Visualization

Heatmap

• Create a heatmap of the top 20 significant genes (without PC1 adjustment).
• Create a heatmap of the top 20 significant genes (with PC1 adjustment).

# Heatmap of top 20 significant genes
top_20_genes <- significant_genes_no_pc1$Gene[1:20]
heatmap_data <- as.matrix(genes_subset[, top_20_genes])

heatmap(heatmap_data, Colv = NA, scale = "row", 
        col = colorRampPalette(c("blue", "white", "red"))(100),
        main = "Expression of Top 20 Significant Genes Without PC1 Adjustment")

top_20_genes_pc1 <- significant_genes_pc1$Gene[1:20]
heatmap_data_pc1 <- as.matrix(genes_subset[, top_20_genes_pc1])

heatmap(heatmap_data_pc1, Colv = NA, scale = "row", 
        col = colorRampPalette(c("blue", "white", "red"))(100),
        main = "Expression of Top 20 Significant Genes With PC1 Adjustment")

Volcano plot

• Calculate log2 fold changes for the whole genes.
• Add the log2 fold changes to results dataframes.

# Calculate log2 fold changes manually
phenotype <- full_dataframe$type

logFC <- apply(expr_norm, 2, function(x) {
  tumor_mean <- mean(x[phenotype != "normal"])
  normal_mean <- mean(x[phenotype == "normal"])
  log2(tumor_mean / normal_mean)
})

# Add logFC to your results dataframes
results_without_pc1$logFC <- logFC[match(results_without_pc1$Gene, names(logFC))]
results_with_pc1$logFC <- logFC[match(results_with_pc1$Gene, names(logFC))]

• Create volcano plots for the results without and with PC1 adjustment.

create_volcano_plot_all <- function(results_df, title) {
  volcano_data <- results_df %>%
    mutate(
      neg_log_pval = -log10(p_value),
      direction = case_when(
        p_value >= 0.05 ~ "Non-sig",
        logFC > 0 ~ "Up",
        logFC <= 0 ~ "Down"
      ),
      significance = ifelse(p_value < 0.05, "Significant", "Non-significant"),
      top_20 = ifelse(p_value < 0.05 & rank(p_value) <= 20, TRUE, FALSE)
    )

  ggplot(volcano_data, aes(x = logFC, y = neg_log_pval)) +
    # Non-significant points
    geom_point(
      data = filter(volcano_data, significance == "Non-significant"),
      aes(color = direction),
      alpha = 0.3,
      size = 2
    ) +
    # Significant points (but not top 20)
    geom_point(
      data = filter(volcano_data, significance == "Significant", !top_20),
      aes(color = direction),
      alpha = 0.6,
      size = 2
    ) +
    # Top 20 points - always green
    geom_point(
      data = filter(volcano_data, top_20),
      color = "green3",  # fixed color
      size = 3,
      shape = 21,        # filled circle with border
      fill = "green3"
    ) +
    # Top 20 labels
    geom_text(
      data = filter(volcano_data, top_20),
      aes(label = Gene),
      color = "green4",  # darker green for text
      vjust = 1.5,
      hjust = 0.5,
      size = 3,
      show.legend = FALSE
    ) +
    scale_color_manual(
      values = c("Down" = "blue", "Up" = "red", "Non-sig" = "grey50"),
      labels = c("Downregulated", "Non-significant", "Upregulated")
    ) +
    geom_vline(xintercept = 0, linetype = "dashed", color = "grey50") +
    geom_hline(yintercept = -log10(0.05), linetype = "dashed", color = "grey50") +
    labs(
      x = "log2 Fold Change (Tumor/Normal)",
      y = "-log10(p-value)",
      title = title,
      color = "Gene Expression"
    ) +
    theme_minimal() +
    theme(
      legend.position = "bottom",
      plot.title = element_text(hjust = 0.5)
    )
}

• Draw the volcano plots.

# Create plots
create_volcano_plot_all(results_without_pc1, "Volcano Plot Without PC1 Adjustment")

create_volcano_plot_all(results_with_pc1, "Volcano Plot With PC1 Adjustment")

Part 3: Annotation

Save the top 20 genes from both analyses to text files.

# For top_20_genes (without PC3 adjustment)
print(top_20_genes)

##  [1] "X208652_at"   "X227004_at"   "X214246_x_at" "X227392_at"   "X200694_s_at"
##  [6] "X238661_at"   "X224297_s_at" "X202974_at"   "X213678_at"   "X213009_s_at"
## [11] "X224471_s_at" "X210550_s_at" "X214825_at"   "X236812_at"   "X237802_at"  
## [16] "X209407_s_at" "X241717_at"   "X228070_at"   "X232275_s_at" "X212448_at"

# Remove first character from each gene name
top_20_genes_trimmed <- sub("^.", "", top_20_genes)  # ^. matches first character

write.table(top_20_genes_trimmed, 
            file = "top20_genes.txt", 
            quote = FALSE,      
            row.names = FALSE,   
            col.names = FALSE)  

# For top_20_genes_pc3 (with PC3 adjustment)
print(top_20_genes_pc1)

##  [1] "X201180_s_at"  "X211778_s_at"  "X218423_x_at"  "X219219_at"   
##  [5] "X201252_at"    "X214381_at"    "X228192_at"    "X1559355_at"  
##  [9] "X219398_at"    "X206272_at"    "X225313_at"    "X203270_at"   
## [13] "X224728_at"    "X213203_at"    "X212690_at"    "X238621_at"   
## [17] "X239667_at"    "X1559546_s_at" "X213009_s_at"  "X221504_s_at"

# Remove first character from each gene name
top_20_genes_pc1_trimmed <- sub("^.", "", top_20_genes_pc1)

write.table(data.frame(Gene = top_20_genes_pc1_trimmed), 
            file = "top20_genes_pc1.txt", 
            quote = FALSE, 
            row.names = FALSE, 
            col.names = FALSE)

Use david tools to convert the top 20 significant gene names to normal gene names.

Normal gene names(no PC1)

Normal gene names(with PC1)

Install and load the hgu133plus2.db package.

if (!require("BiocManager", quietly = TRUE))
    install.packages("BiocManager")
BiocManager::install("hgu133plus2.db") 

library(hgu133plus2.db)

Map the probe IDs to gene symbols using the mapIds function.

gene_symbols <- mapIds(hgu133plus2.db, 
                      keys = top_20_genes_trimmed,
                      column = "SYMBOL",
                      keytype = "PROBEID")

## 'select()' returned 1:1 mapping between keys and columns

print(gene_symbols)

##    208652_at    227004_at  214246_x_at    227392_at  200694_s_at    238661_at 
##     "PPP2CA"      "CDKL5"      "MINK1"      "NISCH"      "DDX24" "MIR124-2HG" 
##  224297_s_at    202974_at    213678_at  213009_s_at  224471_s_at  210550_s_at 
##     "SPTBN4"       "MPP1"   "TMEM151B"     "TRIM37"       "BTRC"    "RASGRF1" 
##    214825_at    236812_at    237802_at  209407_s_at    241717_at    228070_at 
##      "NALF1"      "STMN4"       "XKR4"      "DEAF1"       "MOBP"    "PPP2R5E" 
##  232275_s_at    212448_at 
##     "HS6ST3"     "NEDD4L"

gene_symbols_pc1 <- mapIds(hgu133plus2.db, 
                      keys = top_20_genes_pc1_trimmed,
                      column = "SYMBOL",
                      keytype = "PROBEID")

## 'select()' returned 1:1 mapping between keys and columns

print(gene_symbols_pc1)

##  201180_s_at  211778_s_at  218423_x_at    219219_at    201252_at    214381_at 
##      "GNAI3"      "OVOL2"      "VPS54"    "TMEM160"      "PSMC4" "SEPTIN7P11" 
##    228192_at   1559355_at    219398_at    206272_at    225313_at    203270_at 
##      "UQCC2"      "NXPH2"      "CIDEC"      "RAB4A"    "FAM217B"      "DTYMK" 
##    224728_at    213203_at    212690_at    238621_at    239667_at 1559546_s_at 
##     "ATPAF1"     "SNAPC5"      "DDHD2"       "FMN1"     "SLC3A1"      "SNRPN" 
##  213009_s_at  221504_s_at 
##     "TRIM37"    "ATP6V1H"

Extracting Kegg pathways annotation

• Install and load the enrichR package.
• Perform KEGG pathway enrichment analysis using the enrichR function.
• Specify the KEGG database.

install.packages("enrichR")
library(enrichR)

dbs <- c("KEGG_2021_Human") # Specify KEGG database

• Extract the KEGG results.
• Save the results to a CSV file.

enriched <- enrichr(gene_symbols, dbs)

## Uploading data to Enrichr... Done.
##   Querying KEGG_2021_Human... Done.
## Parsing results... Done.

# Extract KEGG results
kegg_results <- enriched[["KEGG_2021_Human"]]

# Save results
write.csv(kegg_results, "kegg_pathways.csv", row.names = FALSE)
print(kegg_results)

##                                         Term Overlap      P.value
## 1                             Oocyte meiosis   3/129 0.0002757874
## 2             Ubiquitin mediated proteolysis   3/140 0.0003507100
## 3                  mRNA surveillance pathway    2/98 0.0042631106
## 4             Sphingolipid signaling pathway   2/119 0.0062188008
## 5                     AMPK signaling pathway   2/120 0.0063204270
## 6                       Dopaminergic synapse   2/132 0.0075988978
## 7     Adrenergic signaling in cardiomyocytes   2/150 0.0097166465
## 8                    Hippo signaling pathway   2/163 0.0113913007
## 9                             Tight junction   2/169 0.0122042960
## 10                          Circadian rhythm    1/31 0.0305618601
## 11 Aldosterone-regulated sodium reabsorption    1/37 0.0363736593
## 12            Human papillomavirus infection   2/331 0.0426431405
## 13                PI3K-Akt signaling pathway   2/354 0.0481252626
## 14            Glycosaminoglycan biosynthesis    1/53 0.0517104132
## 15                Hedgehog signaling pathway    1/56 0.0545601179
## 16                      Long-term depression    1/60 0.0583470713
## 17                TGF-beta signaling pathway    1/94 0.0899594741
## 18                            Chagas disease   1/102 0.0972497716
## 19                                 Autophagy   1/137 0.1284974931
## 20                       Cellular senescence   1/156 0.1450274169
## 21                               Hepatitis C   1/157 0.1458891094
## 22                     Wnt signaling pathway   1/166 0.1536073032
## 23                            Focal adhesion   1/201 0.1829972755
## 24  Human immunodeficiency virus 1 infection   1/212 0.1920320863
## 25                     Ras signaling pathway   1/232 0.2082163693
## 26                               Shigellosis   1/246 0.2193616046
## 27                               Endocytosis   1/252 0.2240923673
## 28                    MAPK signaling pathway   1/294 0.2564531517
##    Adjusted.P.value Old.P.value Old.Adjusted.P.value Odds.Ratio Combined.Score
## 1        0.00490994           0                    0  27.806723     227.900566
## 2        0.00490994           0                    0  25.559897     203.343059
## 3        0.03539439           0                    0  23.013889     125.604195
## 4        0.03539439           0                    0  18.863248      95.828660
## 5        0.03539439           0                    0  18.702448      94.708609
## 6        0.03546152           0                    0  16.965812      82.788956
## 7        0.03796892           0                    0  14.888889      68.993842
## 8        0.03796892           0                    0  13.677709      61.206452
## 9        0.03796892           0                    0  13.182302      58.080791
## 10       0.08557321           0                    0  35.000000     122.080086
## 11       0.09258750           0                    0  29.157895      96.626651
## 12       0.09950066           0                    0   6.636609      20.937765
## 13       0.10184555           0                    0   6.195707      18.797453
## 14       0.10184555           0                    0  20.170040      59.745598
## 15       0.10184555           0                    0  19.066986      55.455415
## 16       0.10210737           0                    0  17.770740      50.492824
## 17       0.14816855           0                    0  11.254669      27.105700
## 18       0.15127742           0                    0  10.359041      24.141462
## 19       0.18936473           0                    0   7.679567      15.757287
## 20       0.19451881           0                    0   6.731749      12.997879
## 21       0.19451881           0                    0   6.688259      12.874287
## 22       0.19550020           0                    0   6.320574      11.840685
## 23       0.22277929           0                    0   5.205263       8.840015
## 24       0.22403743           0                    0   4.931155       8.136863
## 25       0.23239208           0                    0   4.499658       7.060762
## 26       0.23239208           0                    0   4.239527       6.431506
## 27       0.23239208           0                    0   4.136926       6.187588
## 28       0.25645315           0                    0   3.536375       4.812332
##                  Genes
## 1  PPP2CA;PPP2R5E;BTRC
## 2   NEDD4L;TRIM37;BTRC
## 3       PPP2CA;PPP2R5E
## 4       PPP2CA;PPP2R5E
## 5       PPP2CA;PPP2R5E
## 6       PPP2CA;PPP2R5E
## 7       PPP2CA;PPP2R5E
## 8          PPP2CA;BTRC
## 9        PPP2CA;NEDD4L
## 10                BTRC
## 11              NEDD4L
## 12      PPP2CA;PPP2R5E
## 13      PPP2CA;PPP2R5E
## 14              HS6ST3
## 15                BTRC
## 16              PPP2CA
## 17              PPP2CA
## 18              PPP2CA
## 19              PPP2CA
## 20                BTRC
## 21              PPP2CA
## 22                BTRC
## 23             RASGRF1
## 24                BTRC
## 25             RASGRF1
## 26                BTRC
## 27              NEDD4L
## 28             RASGRF1

NO_PC1

NO_PC1

NO_PC1

enriched_pc1 <- enrichr(gene_symbols_pc1, dbs)

## Uploading data to Enrichr... Done.
##   Querying KEGG_2021_Human... Done.
## Parsing results... Done.

# Extract KEGG results
kegg_results_pc1 <- enriched_pc1[["KEGG_2021_Human"]]

# Save results
write.csv(kegg_results_pc1, "kegg_pathways_pc3.csv", row.names = FALSE)
print(kegg_results_pc1)

##                                                          Term Overlap
## 1                                           Parkinson disease   2/249
## 2                                                  Proteasome    1/46
## 3                                           Cocaine addiction    1/49
## 4                                   Vibrio cholerae infection    1/50
## 5                       Regulation of lipolysis in adipocytes    1/55
## 6                                       Pyrimidine metabolism    1/56
## 7                                        Long-term depression    1/60
## 8                                             Renin secretion    1/69
## 9  Epithelial cell signaling in Helicobacter pylori infection    1/70
## 10                                                  Pertussis    1/76
## 11                                     Gastric acid secretion    1/76
## 12                                     Synaptic vesicle cycle    1/78
## 13                                               Gap junction    1/88
## 14                                          GABAergic synapse    1/89
## 15                                         Morphine addiction    1/91
## 16                                       Rheumatoid arthritis    1/93
## 17                                      Circadian entrainment    1/97
## 18                    Progesterone-mediated oocyte maturation   1/100
## 19                                              Melanogenesis   1/101
## 20                                             Chagas disease   1/102
## 21                           Protein digestion and absorption   1/103
## 22        Parathyroid hormone synthesis, secretion and action   1/106
## 23                                              Toxoplasmosis   1/112
## 24                                       Serotonergic synapse   1/113
## 25                                        Cholinergic synapse   1/113
## 26                       Leukocyte transendothelial migration   1/114
## 27                                      Glutamatergic synapse   1/114
## 28                             Sphingolipid signaling pathway   1/119
## 29             Growth hormone synthesis, secretion and action   1/119
## 30                                        Platelet activation   1/124
## 31                                                   Lysosome   1/128
## 32                                  Relaxin signaling pathway   1/129
## 33                                       Dopaminergic synapse   1/132
## 34                                  Oxidative phosphorylation   1/133
## 35                                   Apelin signaling pathway   1/137
## 36                                 Estrogen signaling pathway   1/137
## 37                             Ubiquitin mediated proteolysis   1/140
## 38                                     Spinocerebellar ataxia   1/143
## 39                       Retrograde endocannabinoid signaling   1/148
## 40                     Adrenergic signaling in cardiomyocytes   1/150
## 41                                                  Phagosome   1/152
## 42                                     mTOR signaling pathway   1/154
## 43                                 Oxytocin signaling pathway   1/154
## 44                                           Cushing syndrome   1/155
## 45                                 cGMP-PKG signaling pathway   1/167
## 46                                               Tuberculosis   1/180
## 47                                              Axon guidance   1/182
## 48                                                 Alcoholism   1/186
## 49                                Chemokine signaling pathway   1/192
## 50                               Epstein-Barr virus infection   1/202
## 51                                     Rap1 signaling pathway   1/210
## 52                   Human immunodeficiency virus 1 infection   1/212
## 53                                     cAMP signaling pathway   1/216
## 54                            Human cytomegalovirus infection   1/225
## 55                                    Chemical carcinogenesis   1/239
## 56                                                Endocytosis   1/252
## 57                                              Prion disease   1/273
## 58                                         Huntington disease   1/306
## 59                             Human papillomavirus infection   1/331
## 60                              Amyotrophic lateral sclerosis   1/364
## 61                                          Alzheimer disease   1/369
## 62                              Pathways of neurodegeneration   1/475
## 63                                         Pathways in cancer   1/531
##       P.value Adjusted.P.value Old.P.value Old.Adjusted.P.value Odds.Ratio
## 1  0.02530959        0.2064179           0                    0   8.876743
## 2  0.04502934        0.2064179           0                    0  23.315789
## 3  0.04789811        0.2064179           0                    0  21.855263
## 4  0.04885255        0.2064179           0                    0  21.408163
## 5  0.05361112        0.2064179           0                    0  19.421053
## 6  0.05456012        0.2064179           0                    0  19.066986
## 7  0.05834707        0.2064179           0                    0  17.770740
## 8  0.06681510        0.2064179           0                    0  15.411765
## 9  0.06775151        0.2064179           0                    0  15.187643
## 10 0.07335128        0.2064179           0                    0  13.968421
## 11 0.07335128        0.2064179           0                    0  13.968421
## 12 0.07521075        0.2064179           0                    0  13.604238
## 13 0.08445508        0.2064179           0                    0  12.034483
## 14 0.08537467        0.2064179           0                    0  11.897129
## 15 0.08721122        0.2064179           0                    0  11.631579
## 16 0.08904426        0.2064179           0                    0  11.377574
## 17 0.09269987        0.2064179           0                    0  10.901316
## 18 0.09543242        0.2064179           0                    0  10.569378
## 19 0.09634153        0.2064179           0                    0  10.463158
## 20 0.09724977        0.2064179           0                    0  10.359041
## 21 0.09815715        0.2064179           0                    0  10.256966
## 22 0.10087408        0.2064179           0                    0   9.962406
## 23 0.10628464        0.2064179           0                    0   9.421053
## 24 0.10718338        0.2064179           0                    0   9.336466
## 25 0.10718338        0.2064179           0                    0   9.336466
## 26 0.10808127        0.2064179           0                    0   9.253377
## 27 0.10808127        0.2064179           0                    0   9.253377
## 28 0.11255786        0.2064179           0                    0   8.859054
## 29 0.11255786        0.2064179           0                    0   8.859054
## 30 0.11701310        0.2064179           0                    0   8.496791
## 31 0.12056198        0.2064179           0                    0   8.227518
## 32 0.12144709        0.2064179           0                    0   8.162829
## 33 0.12409731        0.2064179           0                    0   7.974689
## 34 0.12497903        0.2064179           0                    0   7.913876
## 35 0.12849749        0.2064179           0                    0   7.679567
## 36 0.12849749        0.2064179           0                    0   7.679567
## 37 0.13112751        0.2064179           0                    0   7.512685
## 38 0.13374999        0.2064179           0                    0   7.352854
## 39 0.13810408        0.2064179           0                    0   7.100967
## 40 0.13983989        0.2064179           0                    0   7.004945
## 41 0.14157238        0.2064179           0                    0   6.911467
## 42 0.14330155        0.2064179           0                    0   6.820433
## 43 0.14330155        0.2064179           0                    0   6.820433
## 44 0.14416490        0.2064179           0                    0   6.775803
## 45 0.15446078        0.2162451           0                    0   6.282181
## 46 0.16548182        0.2238105           0                    0   5.822111
## 47 0.16716521        0.2238105           0                    0   5.757197
## 48 0.17052230        0.2238105           0                    0   5.631579
## 49 0.17553385        0.2256864           0                    0   5.453017
## 50 0.18382257        0.2316164           0                    0   5.179104
## 51 0.19039649        0.2321418           0                    0   4.978847
## 52 0.19203209        0.2321418           0                    0   4.931155
## 53 0.19529388        0.2321418           0                    0   4.838433
## 54 0.20258723        0.2363518           0                    0   4.641917
## 55 0.21380775        0.2449071           0                    0   4.365767
## 56 0.22409237        0.2521039           0                    0   4.136926
## 57 0.24043647        0.2657456           0                    0   3.813467
## 58 0.26546094        0.2883455           0                    0   3.395168
## 59 0.28389531        0.3031425           0                    0   3.133971
## 60 0.30755650        0.3212753           0                    0   2.844280
## 61 0.31107607        0.3212753           0                    0   2.804920
## 62 0.38181153        0.3879698           0                    0   2.165889
## 63 0.41633879        0.4163388           0                    0   1.931480
##    Combined.Score       Genes
## 1       32.635986 PSMC4;GNAI3
## 2       72.289232       PSMC4
## 3       66.411133       GNAI3
## 4       64.630147     ATP6V1H
## 5       56.825976       GNAI3
## 6       55.455415       DTYMK
## 7       50.492824       GNAI3
## 8       41.701557       GNAI3
## 9       40.883745     ATP6V1H
## 10      36.492436       GNAI3
## 11      36.492436       GNAI3
## 12      35.200436     ATP6V1H
## 13      29.743651       GNAI3
## 14      29.275335       GNAI3
## 15      28.374333       GNAI3
## 16      27.518048     ATP6V1H
## 17      25.927561       GNAI3
## 18      24.831030       GNAI3
## 19      24.482281       GNAI3
## 20      24.141462       GNAI3
## 21      23.808321      SLC3A1
## 22      22.852586       GNAI3
## 23      21.118557       GNAI3
## 24      20.850327       GNAI3
## 25      20.850327       GNAI3
## 26      20.587577       GNAI3
## 27      20.587577       GNAI3
## 28      19.350725       GNAI3
## 29      19.350725       GNAI3
## 30      18.229604       GNAI3
## 31      17.406064     ATP6V1H
## 32      17.209501       GNAI3
## 33      16.640697       GNAI3
## 34      16.457769     ATP6V1H
## 35      15.757287       GNAI3
## 36      15.757287       GNAI3
## 37      15.262658      TRIM37
## 38      14.792346       PSMC4
## 39      14.058122       GNAI3
## 40      13.780529       GNAI3
## 41      13.511533     ATP6V1H
## 42      13.250766     ATP6V1H
## 43      13.250766       GNAI3
## 44      13.123359       GNAI3
## 45      11.733953       GNAI3
## 46      10.473360     ATP6V1H
## 47      10.298317       GNAI3
## 48       9.961639       GNAI3
## 49       9.487832       GNAI3
## 50       8.772286       PSMC4
## 51       8.258147       GNAI3
## 52       8.136863       GNAI3
## 53       7.902370       GNAI3
## 54       7.411214       GNAI3
## 55       6.734973       GNAI3
## 56       6.187588       RAB4A
## 57       5.435333       PSMC4
## 58       4.502969       PSMC4
## 59       3.946139     ATP6V1H
## 60       3.353681       PSMC4
## 61       3.275355       PSMC4
## 62       2.085379       PSMC4
## 63       1.692471       GNAI3

PC1

PC1

PC1

PC1

• Visualize the KEGG pathway enrichment results.
• Filter significant pathways (p < 0.05) and create a bar plot.

# Filter significant pathways (p < 0.05)
sig_pathways <- kegg_results[kegg_results$Adjusted.P.value < 0.05, ]

ggplot(sig_pathways, 
       aes(x = reorder(Term, -log10(Adjusted.P.value)), 
              y = -log10(Adjusted.P.value))) +
  geom_bar(stat = "identity", fill = "steelblue") +
  coord_flip() +
  labs(title = "KEGG Pathway Enrichment",
       x = "Pathway",
       y = "-log10(Adjusted P-value)") +
  theme_minimal()

# Filter significant pathways (p < 0.05)
sig_pathways_pc1 <- kegg_results_pc1[kegg_results_pc1$Adjusted.P.value < 0.05, ]
ggplot(sig_pathways_pc1, 
       aes(x = reorder(Term, -log10(Adjusted.P.value)), 
              y = -log10(Adjusted.P.value))) +
  geom_bar(stat = "identity", fill = "steelblue") +
  coord_flip() +
  labs(title = "KEGG Pathway Enrichment (PC3 Adjustment)",
       x = "Pathway",
       y = "-log10(Adjusted P-value)") +
  theme_minimal()

Comments on the version with PC1

• Statistically, none of the pathways were significant after adjusting for multiple testing (all adjusted p-value > 0.05).
• Some pathways still show strong signals, with high odds ratios and low raw p-values, suggesting they might be biologically relevant.
• If we’re doing exploratory analysis, it might make sense to relax the threshold a bit (looking at adjusted p-values below 0.15) to catch pathways that could still be worth investigating.